search

tsattler / RansacLib

Template-based implementation of RANSAC and its variants in C++
BSD 3-Clause "New" or "Revised" License
356 stars 48 forks source link

LO-RANSAC: Least squares step before iterative least squares refinement needed? #19

Closed atasoglou closed 1 year ago

atasoglou commented 1 year ago

In the local optimization step of the LO-RANSAC just before the iterative least squares refinement there is again a LSQ fit. Is that really doing anything?

So is this line doing anything? Is this step needed?

Thanks!

tsattler commented 1 year ago

This step is used to polish the most recently found model as to recover as many inliers as possible for the least squares fit. In some cases, this might not be necessary, but in general, I recommend using it.